William S. Charlton

Proliferation Resistance Assessment Methodology for Nuclear Fuel Cycles

A methodology, based on the multiattribute utility analysis, for the assessment of diverse fuel cycles for proliferation resistance was developed. This methodology is intended to allow for the assessment of the effectiveness of safeguards implementation at facilities within a large-scale fuel cycle and allow for the ability to choose technologies based in part on their effectiveness to deter the proliferation of nuclear materials. Fuel cycle facilities under consideration include nuclear reactors, reprocessing facilities, fuel storage facilities, enrichment plants, fuel fabrication plants, uranium conversion plants, and uranium mining and milling operations. The method uses a series of attributes (for example, Department of Energy attractiveness level, weight fraction of even Pu isotopes, measurement uncertainty, etc.) to determine a proliferation resistance measure for each step in a process flow sheet. Each of the attributes has a weighting that determines its importance in the overall assessment. Each attribute also has an associated utility function derived from both expert knowledge and physical characteristics that relates changes in the value of the attribute to its overall effect on the proliferation resistance measure. A method for aggregating proliferation resistance values for each process in a flow sheet into an overall nuclear security measure for the complete cycle was also developed. This method is focused on preventing host nation diversion; however, a similar technique could be used to analyze the risk due to theft by an insider or outsider. This methodology has been applied successfully for example fuel cycles to demonstrate its viability as an assessment methodology and its capability in discriminating diverse fuel cycle options.

Operator declaration verification technique for spent fuel at reprocessing facilities

Abstract A verification technique for use at reprocessing facilities, which integrates existing technologies to strengthen safeguards through the use of environmental monitoring, has been developed at Los Alamos National Laboratory. This technique involves the measurement of isotopic ratios of stable noble fission gases from on-stack emissions during reprocessing of spent fuel using high-precision mass spectrometry. These results are then compared to a database of calculated isotopic ratios using a data analysis method to determine specific fuel parameters (e.g., burnup, fuel type, reactor type, etc.). These inferred parameters can be used to verify operator declarations. The integrated system (mass spectrometry, reactor modeling, and data analysis) has been validated using on-stack measurements during reprocessing of fuel from a US production reactor. These measurements led to an inferred burnup that matched the declared burnup to within 3.9%, suggesting that the current system is sufficient for most safeguards applications. Partial system validation using gas samples from literature measurements of power reactor fuel has been reported elsewhere. This has shown that the technique developed here may have some difficulty distinguishing pressurized water reactor (PWR) from boiling water reactor (BWR) fuel; however, it consistently can distinguish light water reactor (either PWR or BWR) fuels from other reactor fuel types. Future validations will include advanced power reactor fuels (such as breeder reactor fuels) and research reactor fuels as samples become available.

Status of six-group delayed neutron data and relationships between delayed neutron parameters from the macroscopic and microscopic approaches

Work performed in part for an American Nuclear Society Standards Committee Subgroup (ANS 19.9) to assess the status of delayed neutron data is summarized. Recent measurements of delayed neutron emission conducted at Texas A and M University are also described. During the last 10 yr, there have been advances in nuclear data libraries (e.g., improved fission product yields) that make it possible to quantitatively predict delayed neutron emission from basic data. The six-group delayed neutron data available in the literature from both macroscopic level experiments and microscopic level calculations for several actinide isotopes are compared. Results are also presented from recent experimental measurements of delayed neutron emission and delineates some of the relationships between these measurements and microscopic level predictions. For example, from the experimental measurements, Keepin`s delayed neutron group 1 is shown to correspond mainly to a single isotope. {sup 87}Br, as expected from microscopic level theory, and Keepin`s group 2 is shown to correspond to primarily two separate isotopes. {sup 137}I and {sup 88}Br. In the future, it may be useful to use properties of specific isotopes to replace Keepin`s delayed neutron groups 1, 2, 3, and 4 for prescribing delayed neutron data for actinides.

Comparisons of Calculated and Measured 237Np, 241Am, and 243Am Concentrations as a Function of the 240Pu/239Pu Isotopic Ratio in Spent Fuel

The Los Alamos National Laboratory (LANL) has developed a system for determining 237 Np, 241 Am, and 243 Am concentrations in spent fuel from measurements of the 240 Pu/ 239 Pu isotopic ratio using calculations performed with the HELIOS lattice-physics code. Benchmark calculations for several pressurized water reactors (PWRs) were performed and compared to measured values from the literature for fuels with burnups ranging from 0 to 50000 MWd/tonne U. A direct correlation can be found between the 240 Pu/ 239 Pu isotopic ratio and the higher-actinide concentrations for each fuel type. Comparisons of calculated with measured values suggests that the LANL technique would yield 237 Np and 241 Am concentrations within±5% and 243 Am concentrations within ±15% for PWRs. Expanding this system for all reprocessing applications will require more measured data (especially for boiling water reactors and VVER-type reactors), but the existing results show a marked improvement over the previous ORIGEN calculations. Also, a better determination of the 243 Am concentrations may support a greater confidence in the calculated results or suggest an alteration to the existing nuclear data. The present state ofthese neutronics calculations suggests that the technology exists to reduce the need for direct measurement of the 237 Np, 241 Am, and 243 Am concentrations in spent fuel.

Benchmarking of the Los Alamos neutron production rate code SOURCES-3A

Abstract The neutron production rate code SOURCES-3A was benchmarked using various experimental measurements from the literature. These experiments included thick-target yield measurements from Li, Be, B, C, O, F, Mg, Al, Si, UO 2 , and UC. Several integrated experiments (PuBe 13 and UO 2 F 2 homogeneous problems, Po–B 4 C and Po–Be interface problems, and Al 2 O 3 and SiO 2 beam problems) were also modeled, testing all the geometry characteristics of the SOURCES-3A code. The SOURCES-3A calculations were compared with the experimental results and good agreement was found in all cases. These benchmarks have shown that SOURCES-3A spectra and magnitude calculations are accurate to within ±18% for even the most complex problems.

Preliminary results of nuclear fluorescence imaging of alpha and beta emitting sources

The preliminary results from a series of nuclear fluorescence imaging experiments using a variety of radioactive sources and shielding are given. These experiments were done as part of a proof of concept to determine if nuclear fluorescence imaging could be used as a safeguards measurements tool or for nuclear warhead verification for nuclear arms control treaties such as the New Strategic Arms Reduction Treaty and the Fissile Material Cut-Off Treaty. An off-the-shelf Princeton Instruments charged coupled device camera system was used to image the emission of fluorescence photons from the de-excitation of nitrogen molecules in air that have been excited by ionizing radiation. The fluorescence emissions are primarily in the near ultraviolet range; between the wavelengths of 300 and 400 nm. Fluorescent imaging techniques are currently being investigated in a number of applications. A French research team has successfully demonstrated this concept for remote imaging of alpha contamination. It has also been shown that the phenomenon can be seen through translucent materials and that alpha radiation can be seen in the presence of large gamma backgrounds. Additionally, fluorescence telescopes and satellites utilize the de-excitation of nitrogen molecules to observe cosmic ray showers in the atmosphere. In cosmic ray shower detection, electrons are the main contributor to the excitation of the of nitrogen molecules in air. The experiments presented in this paper were designed to determine if the imaging system could observe beta emitting sources, differentiate between beta emitters and alpha emitting materials such as uranium oxide and uranium metal, and to further investigate the phenomenon through translucent and non-translucent materials. The initial results show that differentiation can be made between beta and alpha emitting sources and that the device can observe the phenomenon through very thin non-transparent material. Additionally, information is given on the detection of the fluorescent photons through translucent materials. Camera images, analysis, and results of the initial laboratory experiments are presented.

Development of Self-Interrogation Neutron Resonance Densitometry to Improve Detection of Possible Diversions for PWR Spent Fuel Assemblies

A new nondestructive assay technique called self-interrogation neutron resonance densitometry (SINRD) is currently being developed at Los Alamos National Laboratory to improve existing nuclear safeguards and material accountability measurements for light water reactor fuel assemblies. The viability of using SINRD to improve the detection of possible diversion scenarios for pressurized water reactor 17 × 17 spent low-enriched uranium (LEU) and mixed oxide (MOX) fuel assemblies was investigated via Monte Carlo N-Particle eXtended transport code (MCNPX) simulations. The following capabilities were assessed: (a) verification of the burnup of a spent fuel assembly, (b) ability to distinguish fresh and one-cycle spent MOX fuel from three- and four-cycle spent LEU fuel, and (c) sensitivity and penetrability to the removal of fuel pins. SINRD utilizes 244Cm spontaneous-fission neutrons to self-interrogate the spent fuel pins. The amount of resonance absorption of these neutrons in the fuel can be quantified using a set of fission chambers (FCs) placed adjacent to the assembly. The sensitivity of SINRD is based on using the same fissile materials in the FCs as are present in the fuel because the effect of resonance absorption lines in the transmitted flux is amplified by the corresponding (n, f) reaction peaks in the FC. SINRD requires calibration with a reference assembly of similar geometry in a similar measurement configuration with the same surrounding moderator as the spent fuel assemblies. However, this densitometry method uses ratios of different detectors so that several systematic errors related to calibration and positioning cancel in the ratios.

Development of Self-Interrogation Neutron Resonance Densitometry to Quantify the Fissile Content in PWR Spent LEU and MOX Assemblies

Abstract A new nondestructive assay technique called self-interrogation neutron resonance densitometry (SINRD) is currently being developed at Los Alamos National Laboratory to improve existing nuclear safeguards and material accountability measurements for light water reactor fuel assemblies. The viability of using SINRD to quantify the fissile content (235U and 239Pu) in pressurized water reactor 17 × 17 spent low-enriched uranium and mixed-oxide fuel assemblies in water was investigated via Monte Carlo N-particle extended transport code simulations. SINRD utilizes 244Cm spontaneous fission neutrons to self-interrogate the fuel pins. The amount of resonance absorption of these neutrons in the fuel can be quantified using 235U and 239Pu fission chambers placed adjacent to the assembly. The sensitivity of this technique is based on using the same fissile materials in the fission chambers as are present in the fuel because the effect of resonance absorption lines in the transmitted flux is amplified by the corresponding (n,f) reaction peaks in the fission chamber. SINRD requires calibration with a reference assembly of similar geometry. However, this densitometry method uses ratios of different fission chambers so that most systematic errors related to calibration and positioning cancel in the ratios.

Comparisons of HELIOS, ORIGEN2, and Monteburns Calculated 241 Am and 243 Am Concentrations to Measured Values for PWR, BWR, and VVER Spent Fuel

A study was performed at Los Alamos National Laboratory to explore the accuracy of several reactor analysis codes in calculating 241 Am and 243Am concentrations in light water reactor spent fuel. Calculated higher-actinide concentrations were compared to measured values from the literature for three reactor fuels. The fuel samples were taken from the Mihama Unit 3 pressurized water reactor, the Garigliano boiling water reactor, and a VVER-440. The 241Am and 243Am concentrations were calculated using the HELIOS-1.4 lattice-physics code, the ORIGEN2 burnup code, and a linked MCNP/ORIGEN2 code named Monteburns 3.01. Comparisons were made between the calculated and measured values. It was determined that all codes performed consistently well for the Mihama Unit 3 measurements (within ±5% for 241Am and ±20% for 243Am) and the Garigliano measurements (within ±12% for 241 Am and ±20% for 243Am). It was determined that the ORIGEN2 pressurized water reactor libraries are insufficient for the VVER-440 measurements. The HELIOS and MONTEBURNS codes both demonstrated good ability to calculate these isotopes for VVER-440 fuel (±10% for 241Am and ±12% for 243Am). The accuracies of these codes and the associated radiochemical measurements of these higher-actinide isotopes may be insufficient for safeguards and fuel management purposes; thus, development of new methods and modification to existing data libraries may be necessary in order to enable cost-effective safeguarding of these higher-actinide materials.

Predicting Concrete Roadway Contribution to Gamma-Ray Background in Radiation Portal Monitor Systems

Abstract The collapse of the Soviet Union ushered in an era of interest in the security of the radiological and nuclear material holdings of the Russian Federation and other countries of the Former Soviet Union. Additionally, the increasing sophistication of international criminal and terrorist organizations highlighted the need to secure these materials and prevent them from being smuggled from their point of origin and across international boundaries. To combat the growing threat of radiological and nuclear smuggling, radiation portal monitors (RPMs) are deployed at ports of entry (POEs) around the world to passively detect gamma and neutron radiation signatures from cargo and pedestrian traffic. In some locations, RPMs are reporting abnormally high gamma-ray background count rates, a situation that has been attributed, in part, to the building materials surrounding the RPMs. The primary objective of this work was to determine the impact of different types of concrete on the gamma-ray background readings in a particular RPM. Secondary objectives include developing an adaptable model to estimate the gamma-ray background contribution from any composition of concrete in any RPM configuration and determining the elemental composition of different concrete samples through neutron activation analysis (NAA) techniques. The specific activities of 40K and isotopes from the 238U and 232Th decay series were determined with a high-purity germanium detector and computer-generated calibration files. Through NAA, 34 elemental compositions were determined for six concrete samples from three different parent slabs. The total weight percentages determined were 84% to 100% of the total mass of the samples. The Monte Carlo N-Particle (MCNP) transport code was used to simulate the RPM response to the different concrete slabs. The MCNP model was validated by comparing actual and simulated detector responses to 137Cs check sources of varying strengths. For all validation cases, the MCNP estimates were 6% to 16% less than the value obtained from the actual RPM data. This work shows that it is possible to estimate the gamma-ray response of an RPM to the underlying concrete roadway. Knowing the amount of this contribution will allow RPM customers to choose suitable foundation materials before installation and accurately set alarm thresholds. This could ultimately increase the ability of RPMs to detect radiation at POEs, thereby increasing the probability of a seizure of smuggled radiological and nuclear materials.